Do whole-school reform models boost student performance: Evidence from New York City

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Public Administration


John Yinger


Performance improvement, School Development Program, More Effective Schools, Success for All, Reform, New York City

Subject Categories

Education | Educational Administration and Supervision | Public Administration | Public Affairs, Public Policy and Public Administration | Social and Behavioral Sciences


Whole-school reform models are school-level interventions that address multiple aspects of school operations and are intended to improve student performance. This dissertation examines three specific whole-school reform models: the School Development Program, More Effective Schools and Success for All. The primary purpose of the study is to determine the impacts that a school's decision to adopt one of these whole-school reform models has on student academic performance. In addition to reviewing evidence provided by existing evaluations, the dissertation presents a quasi-experimental evaluation of whole-school reform efforts undertaken in New York City during the mid-1990s. The study sample consists of 47 New York City elementary schools that adopted one of the three whole-school reform models, as well as 42 randomly selected comparison schools. Individual-level panel data on students from each school in the study sample are used to estimate model impacts on reading and math performance. The estimation procedures used, including difference-in-differences-in-differences and instrumental variables, pay careful attention to potential biases created by the self-selected nature of the treatment group. The study finds that the School Development Program and More Effective Schools had small, positive impacts on student performance in both math and reading. Students who were exposed to the School Development Program beginning in first grade, by fifth grade show average reading scores that were 0.24 standard deviations higher and math scores that were 0.21 standard deviations higher than we might have expected in the absence of model adoption. Students exposed to More Effective Schools beginning in first or second grade, by fifth grade show reading scores that are 0.31 standard deviations higher and math scores that are 0.35 standard deviations higher than we would have expected in absence of the model. However, estimated impacts of More Effective Schools several years after adoption are smaller and statistically insignificant suggesting that improvements are not maintained. Statistically significant, positive impacts were not found for Success for All. Implications of the analysis for policy and future research are detailed in the concluding chapter.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.